Software development is constantly evolving. Since the 90s, five major waves of software—mainframe, mini, client/server, 3 tier or web, and cloud (better known as SaaS)—created massive disruption, allowing new companies to challenge the old guard and add enormous value. Companies like SAP and Oracle did so with the client/server wave, Siebel and i2 with the web wave, and Salesforce, Success Factors, and many others, with the SaaS wave.
Over the last decade, SaaS in India has gone from being just a new entrant to a forceful game changer. A recent McKinsey and SaaSBoomi report predicts that in India, SaaS will be a $1 trillion market value opportunity by 2030—a staggering figure considering that India’s total public market capitalization is roughly around $3.5 trillion. In addition to a fast-maturing startup ecosystem, the opportunity has been aided by the advent of cloud software. It makes software accessible to smaller buyers, enabling Indian SaaS companies to sell remotely, and utilize cost advantages in development, sales, and customer success.
The latest disruption is being led by “intelligent software” or AI-first software. This category of software uses sophisticated learning models that work on underlying data sets to solve focused business problems. It is probabilistic instead of deterministic, and helps with judgment. To explain, it helps answer what one should do instead of increasing productivity of tasks only after the ‘what’ is answered. Consequently, AI-first software can have a much larger impact on making enterprises faster and efficient.
For example, Signzy helps large financial institutions detect identity fraud. Instead of using a set of static rules to identify a fake driver’s license, the company’s solution does it intelligently, with models updating themselves as they are fed with more data. As a result, banks are spared from constantly identifying new fraud patterns, training a large workforce to identify them, and altering workflow for error logging, exception handling, and so on.
AI solutions fundamentally differ from traditional SaaS, leading to category creation. Code is to SaaS, what data is to AI. In addition to coding skills—primary skills needed for traditional software development—building AI solutions require data engineering, data science, and modeling skills. The inherent separation of data and software disappears with AI, as data determine or ‘write’ the software. Therefore, the stack itself, and skills required to build AI solutions, are distinctly different from those required for traditional SaaS applications.
What deserves mention is that instead of just being a “feature” in existing applications, AI will help produce a new class of applications and create the next enterprise disruption, similar to the impact of cloud at the turn of the 21st Century. Estimates suggest that by 2030, AI infrastructure, applications, and services will become a market worth more than $800B annually.
India opportunity
Unlike previous waves, where India traditionally lagged behind as a provider, the country is well positioned to claim its fair share of the AI pie. A thriving startup ecosystem, the experience of building globally successful SaaS companies, and a large data and analytics talent pool can help India generate a market value of $500 billion from AI applications and services by 2030.
While opportunities exist throughout the stack, we believe India’s opportunities will lie primarily in applications and services. Success factors for applications, application APIs, and services are similar to what India has explored over the last decade in SaaS. We also believe the MLOps space may offer a significant opportunity, though potentially more challenging when compared to applications, APIs, and services. Furthermore, adoption of AI across a variety of sectors can potentially have significant social impact.
In India, a fair number of green shoots are already emerging in the space of AI-first SaaS. GTMBuddy, for example, is recommending appropriate content to salespeople as they interact with their prospects. Yellow.ai is enabling large enterprises to support their customers through a human-like chatbot, and Uniphore is automating call transcription and analysis. Many of these companies have even raised significant capital from leading global funds.
For India, seizing this opportunity comes with its share of challenges including access to senior data science and product talent. While the country has abundant developers and analytics professionals, senior talent is difficult to find. Also, lack of meaningful research in AI puts India at a disadvantage when compared to its global peers. Finally, data is the lifeblood of AI. Many countries have taken initiatives at the national level to create and curate clean data and India has commenced efforts in this direction as well.
The size of the AI prize is too large to ignore, and both the government and industry can work together to mitigate the challenges to AI adoption. Global examples suggest that well-designed policy measures can have a positive impact on a country’s competitiveness in AI. In alignment, as we set the stage to explore promising possibilities, we look forward to some of the supportive regulatory measures, currently underway that are likely to be a game changer for India.
Alok Goyal, is Partner, Stellaris Venture Partners and Ruchira Shukla, Head, South Asia, Disruptive Technologies – Direct Equity and VC Funds of International Finance Corporation (IFC)
Published on
February 14, 2022
